Studies of T cell activation of normal murine T cell clones of the TH1 type have shown that two signals are required to stimulate the cells to make interleukin-2 (IL-2) and divide. One signal is given through the antigen- specific receptor that is uniquely expressed on each clone. The other signal is called costimulation and is delivered through a different receptor that has not yet been characterized. Signal 1 in the absence of signal 2 induces a state of hyporesponsiveness in the T cell clone known as clonal anergy. In this state, the cell fails to make significant amounts of IL-2 when restimulated with both signals 1 and 2. During the past year we have made progress in two areas. First, we have shown that the anergic state can be reversed in T cell clones by stimulating the cells with IL-2. This reversal occurs in all the cells and affects all impaired lymphokine production. Anergy was also found to dissipate spontaneously, although much more slowly, in the absence of IL-2 stimulation. Finally, we showed that a partial state of anergy could be induced by normal activation with antigen and APC provided that IL-2 and possibly other factors were removed from the culture medium by washing the cells. The results demonstrate that the anergic state is not a permanent change in the TH1 cell. They also indicate that anergy induction might be a consequence of the inability of the cell to divide following stimulation. Thus, costimulation may only be required in normal activation to produce IL-2, which in turn prevents anergy by driving the cell through repeated rounds of division. In a separate set of experiments, we demonstrated that costimulation is required in order to make murine TH1 clones competent to proliferate in response to interleukin-4. Induction of anergy in the cells blocked this ability to become competent in parallel with the impairment of the cells to make IL-2; however, IL-2 stimulation was not the only signal required to gain competence, because addition of IL-2 to a T cell-receptor occupancy signal, mediated by concanavalin A, only allowed the cells to become partially competent to respond to IL-4. We concluded from these experiments that costimulatory signals contribute directly to the attainment of full competence to respond to IL-4.